Introduction

There is scarce evidence regarding the impact of capsule endoscopy (CE) training programs.

Aims & Methods

We aimed to develop and validate a simple capsule endoscopy (CE) training assessment tool, the Capsule Endoscopy Training Assessment (CETA), and prospectively use it to analyze the learning progression achieved by participants in our CE training program.
Over a 3-year period, all participants in our CE training program completed pretraining and posttraining CETA, ranging between 0-100%, and encompassing theoretical questions and interpretation of segmented CE videos. We compared the mean differences in overall, theoretical, and practical pretraining-posttraining CETA and assessed the influence of previous endoscopic experience (upper gastrointestinal endoscopy [UGE], colonoscopy, device-assisted enteroscopy [DAE] and CE) using generalized linear models.

Results

Fifty-seven participants were included. After training, there was a significant increase in participants’ overall (mean difference, 26.3; 95% confidence interval [CI], 20.70 to 31.83), theoretical (mean difference, 27.2; 95% CI, 19.81 to 34.57), and practical (mean difference, 25.9; 95% CI, 20.09 to 31.63) CETA components. Compared to those without experience, participants with previous experience with these techniques had higher overall pretraining CETA (UGE, mean difference, 15.8; 95% CI, 5.79 to 25.75; colonoscopy, mean difference, 12.5; 95% CI, 2.35 to 22.65; DAE, mean difference, 14.2; 95% CI, 3.92 to 24.54, CE, mean difference, 16.3; 95% CI, 8.72 to 23.93), lower mean difference between overall pretraining and posttraining CETA (UGE, rate ratio, 0.76; 95% CI, 0.63 to 0.91; colonoscopy, (rate ratio, 0.80; 95% CI, 0.67 to 0.9; DAE, (rate ratio, 0.84; 95% CI, 0.73 to 0.97; CE, rate ratio, 0.81; 95% CI, 0.72 to 0.92, respectively) but not significantly different overall posttraining CETA.

Conclusion

CETA is a valid and useful tool in assessing the learning progression achieved by participants after CE training program. We demonstrated a significant improvement in participants’ CETA after training, being the least experienced participants in endoscopic procedures who benefited the most from CE training.

Introduction

Capsule endoscopy (CE) is an important non-invasive diagnostic modality for the evaluation of small bowel and colon. However, in patients with gastrointestinal tract anatomy altered by surgical interventions, there can be a higher risk of incomplete examination of small bowel and colon.

Aims & Methods

The aim of our study was to evaluate the rate of complete examination and diagnostic yield of CE, in patients with previous gastrointestinal surgery, as well as the impact of endoscopic placement of CE using the AdvanCE device.
Retrospective, cohort study, including patients submitted to CE of the small bowel or colon, detailing patients with previous gastrointestinal surgeries and with endoscopic placement of CE.

Results

A total of 642 patients was included, 17 with previous gastrointestinal surgery (2.6%), from which 9 with gastric bypass and Y-de-Roux (with 3 having esophagojejunostomy and 6 gastrojejunostomy), 3 patients with Bilroth II, 2 with Bilroth I e 3 patients with sleeve gastrectomy.
Endoscopic placement of CE was performed at the discretion of the responsible gastroenterologist and occurred in 8 patients (1.2%). It was 13.3 times more likely to be performed in patients with previous gastrointestinal surgery (11.8% vs 1.0%, p=0.017).
In patients without endoscopic placement of CE, the rate of complete examination of the gastrointestinal tract, adequate bowel preparation and diagnostic yield was comparable between those with and without previous gastrointestinal surgery: 86.7% vs 85.3% (p=1.000), 73.3% vs 78.1% (p=0.751) e 40% vs 38% (p=0.875), respectively.
In patients with gastrointestinal surgery, the rate of complete examination of the gastrointestinal tract was comparable between patients with and without endoscopic placement of CE: 50.0% vs 86.7% (p=0.331). The rate of complete examination of the gastrointestinal tract, adequate bowel preparation and diagnostic yield was not influenced by the different types of surgery, considering patients with gastric bypass vs Y-de-Roux vs Bilroth II: 75.0% vs 80.0% vs 66.7% (p=0.915), 50.0% vs 60.0% vs 66.7% (p=0.902) e 25.0% vs 20.0% vs 66.7% (p=0.363), respectively.

Conclusion

CE is effective, even in patients with previous gastrointestinal surgery, with diagnostic yield and rate of complete examination of the gastrointestinal tract similar to other patients, regardless of the type of surgery. Endoscopic placement of CE with AdvanCE device does not seem to improve outcomes in patients with previous gastrointestinal surgery.

Introduction

Capsule Endoscopy (CE) offers a minimally invasive method for comprehensive visualization of the gastrointestinal (GI) tract. Protruding lesions are relatively common findings in CE, with variable incidence and diverse pathological significance. Accurate detection of these lesions is crucial but remains challenging due to the time-consuming nature of CE interpretation and susceptibility to human error. While artificial intelligence (AI) presents a promising solution, real-world evidence evaluating its clinical performance is still scarce. To assess the diagnostic performance of a deep learning convolutional neural network (CNN) for the detection of protruding lesions in complete CE videos under real-world clinical conditions.

Aims & Methods

Our group conducted a prospective, multicenter validation study comparing AI-assisted CE reading to standard, conventional interpretation. A total of 211 CE videos, obtained using three different CE systems across seven centers in Portugal, Spain, Brazil, and the United States, were analyzed. Following initial standard readings, an independent expert conducted AI-assisted readings using a CNN specifically trained to detect small bowel protruding lesions. Discrepancies between conventional and AI-assisted reports were adjudicated by a third-party expert from an independent center to establish the reference standard. Diagnostic performance was assessed using sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUC-ROC).

Results

AI-assisted reading detected 106 confirmed protruding lesions, compared to 42 identified through conventional reading. Sensitivity was significantly higher with AI-assisted reading (95.9% vs. 64.6%), although specificity was lower (78.7% vs. 96.1%). Overall accuracy was relatively lower with AI-assisted reading (81.2% vs. 91.5%). The AUC-ROC was 0.86 for AI-assisted and 0.80 for standard reading. Additionally, AI-assisted reading reduced the mean examination time to 316 seconds per video.

Conclusion

AI-assisted interpretation of complete CE videos significantly enhances the detection of protruding lesions compared to conventional reading, although at the expense of slightly reduced specificity. The inclusion of videos from multiple CE systems addresses interoperability challenges, while the participation of centers from four countries across two continents supports the model’s generalizability and real-world clinical applicability.

Introduction

Small-bowel (SB) inflammatory activity in Crohn's Disease (CD) is variable among patients. Capsule Enteroscopy (CE) allows for visualization of the entire SB, which may be divided into three parts, known as tertiles.

Aims & Methods

The aim of this study was to determine the association between SB inflammatory involvement in CD of the first and second tertiles and the impact on the clinical outcome. A total of 53 treatment-naive patients with SB CD submitted to CE as part of the diagnosis workup were selected. Data regarding SB inflammatory activity quantified by the Lewis Score (LS; < 135, normal; 135–790, mild; ≥790 moderate-to-severe) were collected. Individuals with any colonic involvement, as well as stricturing or fistulizing phenotypes, were excluded from the analysis.

Results

Thirty-six (67.9%) patients displayed inflammatory activity in the first and/or second tertile together with third tertile involvement (Proximal+T3 group). Seventeen (32.1%) had inflammation in the third tertile only (T3 group). The LS accounts for the SB tertile with the highest degree of inflammation, yet we found that individuals in the Proximal+T3 group exhibited higher inflammatory scores (p=0.031). We performed a subgroup analysis for those with mild inflammatory activity. There were no differences regarding age or sex distribution, smoking habits, frequency in relatives with CD, perianal disease, extra-intestinal manifestations, or laboratory values. Notably, we found that patients with mild disease in the Proximal+T3 group initiated biologic drugs more often in the five years following the initial diagnosis (p=0.036).

Conclusion

This study shows that involvement of the first and/or second tertiles of the SB is significantly associated with more severe inflammatory activity overall. Additionally, in those with mild disease and proximal inflammation the need for biologics occurred more frequently. This suggests that early detection of proximal SB CD using CE may contribute to timely treatment.

Introduction

The most important factors for a successful colon capsule endoscopy (CCE) study are the quality of bowel preparation and the capsule excretion during battery life. Incomplete colonoscopy is one of the main indications for CCE.

Aims & Methods

The aim of this study was to analyze clinical and demographic factors for complete CCE after incomplete colonoscopy.
A retrospective single-center study was conducted, including patients consecutively submitted to CCE with previous incomplete colonoscopy. Complete examination was defined as visualization of hemorrhoidal plexuses during battery time. Demographic (gender and age) and clinical data (diabetes mellitus, hypertension, smoking, BMI, hypothyroidism, constipation, depression, opioid therapy, psychotropic medication and previous abdominal surgery) were collected.

Results

197 patients were analyzed, with a mean age of 67 years, 71.6% were women. The main reasons for incomplete colonoscopy were impassable angulation (54.8%) and irreducible loop (40.1%).
CCE was incomplete in 32.5% (n=64) of cases. Bowel preparation was inadequate in 30.5% (n=60).
Constipation was a risk factor for incomplete CCE study (OR 2.962; p=0.021, 95% CI 1.178-7.264), as well as obesity (OR 4.407; p=0.020, 95% CI 1.263-15.373).
The clinical indication for the examination did not show a statistically significant relationship with the rate of incomplete examinations (p=0.603). There were no statistically significant differences between the percentage of incomplete CCE in patients with incomplete colonoscopy due to impassable angulation versus irreducible loop (33.3% vs 33.3%, p=1.000).

Conclusion

Constipation and obesity are factors associated with a higher risk of incomplete CCE study. In these patients, intensification or alteration of the bowel preparation protocols used may be considered.

Introduction

Double-headed capsules enable the non-invasive one-step endoscopic evaluation of both the small bowel and colon (pan-intestinal capsule endoscopy, PCE).

Aims & Methods

We aim to evaluate the diagnostic accuracy and safety of PCE vs. the current standard of care (conventional colonoscopy) as the first-line examination in patients with iron deficiency anaemia (IDA) and/or overt bleeding after a non-diagnostic upper endoscopy (suspected mid or lower gastrointestinal bleeding, MLGIB).
We conducted a prospective, single-centre, single-blinded cohort study including consecutive patients with suspected MLGIB. All patients were submitted to PCE, followed by same-day colonoscopy under deep sedation. The physicians performing PCE and conventional colonoscopies were independent and blinded to each other’s observations. The diagnostic yield for potentially haemorrhagic lesions (PHL) was compared for efficacy analysis, and adverse events were assessed for both examinations. The gold standard diagnosis was defined by consensus panel. This trial has been registered in ClinicalTrials.gov (ID NCT04782986).

Results

We included 100 consecutive patients with suspected MLGIB, mean age 66.5 (18-92) years, 65% female gender, presenting with IDA (81%) or overt bleeding (19%). Mean haemoglobin level was 8.4 (4.1-12.8) g/dL and 40% were inpatients. For PCE, the PillCam Crohn’s® capsule was used in 88 patients, and PillCam COLON® 2 in the remaining. Patients with overt bleeding performed PCE within 2 weeks in 70% of cases, 15% in the first 72h after the bleeding episode. PCE was complete in 76% of cases, vs. colonoscopy 95% (p<0.01). The quality of bowel preparation was adequate in 72% of PCE vs. 85% of colonoscopies (p=0.012). Overall, PHL were diagnosed, based on the combination of PCE + colonoscopy, in 46% of patients, corresponding to small bowel and/or colon angioectasias in more than two-thirds (69.6%) of cases. PHL were detected by both examinations in 13/46 (28.3%), only by colonoscopy in 3/46 (6.5%), and only by PCE in 30/46 (65.2%) of cases (p<0.01). Overall, colonoscopy detected PHL in 23% of the entire cohort, which means 50% (23/46) of all patients with PHL according to the consensus panel, while for PCE the overall diagnostic yield was 44%, meaning 95.7% (44/46) of all PHL (p<0.01). Considering the gold standard diagnosis for PHL, colonoscopy had a sensitivity of 23/46=50%, specificity 54/54=100%, positive predictive value (PPV) 23/23=100% and negative predictive value (NPV) 54/77=70,1%, while for PCE the sensitivity was 44/46=95.7%, specificity 54/54=100%, PPV 44/44=100% and NPV 54/56=96.4%. Thirty-one percent of patients had PHL accessible to ileocolonoscopy; in those cases, the PCE was able to make the diagnosis in 29/31 (93.5%) vs. conventional colonoscopy in 23/31 (74.2%) of cases, p<0.01. Regarding safety, no significant adverse events occurred with PCE, vs. 2% (cardiopulmonary) with conventional colonoscopy.

Conclusion

PCE is safe and more effective than conventional colonoscopy in identifying PHL both in the small bowel and the colon, and could be considered as a first-line examination in patients with IDA or overt bleeding after a non-diagnostic upper endoscopy. In more than half of patients with suspected MLGIB, PCE does not detect any PHL either in the small bowel or the colon, dismissing any further invasive endoscopic investigations.

Disclosure

Medtronic provided financial support to the acquisition of some of the capsules which were used in the trial